From: Eric Sandeen Subject: Re: "data=writeback" and TRIM don't get along Date: Wed, 07 Apr 2010 23:47:23 -0500 Message-ID: <4BBD5FDB.9010100@redhat.com> References: <4BBD285B.9000603@gmail.com> <4BBD2FDF.4040407@redhat.com> <4BBD3365.90306@gmail.com> <4BBD5740.4070101@redhat.com> <4BBD5D90.4090203@redhat.com> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 7bit Cc: linux-ext4@vger.kernel.org To: Nebojsa Trpkovic Return-path: Received: from mx1.redhat.com ([209.132.183.28]:58397 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750830Ab0DHErj (ORCPT ); Thu, 8 Apr 2010 00:47:39 -0400 In-Reply-To: <4BBD5D90.4090203@redhat.com> Sender: linux-ext4-owner@vger.kernel.org List-ID: Eric Sandeen wrote: > Eric Sandeen wrote: >> I'll have to think about the right way to do this... it seems pretty >> convoluted to me right now. >> > Something like this probably works, but I really REALLY would not test > it on an important filesystem. :) > > I'm not sure it's a good idea to discard it before returning it > to the prealloc pool, because it may well get re-used again > quickly.... not sure if that's helpful. (now I'm really talking to myself, but scratch that bit - ext4_mb_return_to_preallocation is pretty much a no-op) -Eric > Just a note, I think eventually we may move to more of a batch discard > in the background, because these little discards are actually quite > inefficient on the hardware we've tested so far. > > -Eric > > p.s. really. Don't test this with important data. I haven't tested > it at all yet. > > Index: linux-2.6/fs/ext4/mballoc.c > =================================================================== > --- linux-2.6.orig/fs/ext4/mballoc.c > +++ linux-2.6/fs/ext4/mballoc.c > @@ -4602,6 +4606,8 @@ do_more: > mb_clear_bits(bitmap_bh->b_data, bit, count); > ext4_mb_free_metadata(handle, &e4b, new_entry); > } else { > + ext4_fsblk_t discard_block; > + > /* need to update group_info->bb_free and bitmap > * with group lock held. generate_buddy look at > * them with group lock_held > @@ -4609,6 +4615,11 @@ do_more: > ext4_lock_group(sb, block_group); > mb_clear_bits(bitmap_bh->b_data, bit, count); > mb_free_blocks(inode, &e4b, bit, count); > + discard_block = bit + > + ext4_group_first_block_no(sb, block_group); > + trace_ext4_discard_blocks(sb, > + (unsigned long long)discard_block, count); > + sb_issue_discard(sb, discard_block, count); > ext4_mb_return_to_preallocation(inode, &e4b, block, count); > } > > > -- > To unsubscribe from this list: send the line "unsubscribe linux-ext4" in > the body of a message to majordomo@vger.kernel.org > More majordomo info at http://vger.kernel.org/majordomo-info.html